
import numpy as np #для матричных операций
np.set_printoptions(precision=3, suppress=True)
import pandas as pd #для загрузки данных
import matplotlib.pyplot as plt #для графиков
import seaborn as sns #для отрисовки графиков


from sklearn.preprocessing import MinMaxScaler, LabelEncoder #
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error, r2_score, mean_absolute_error
from sklearn.neighbors import KNeighborsRegressor
from sklearn.svm import SVR
from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor
from sklearn.tree import DecisionTreeRegressor
from sklearn.model_selection import GridSearchCV
from pandas.plotting import scatter_matrix


import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers


df_bp = pd.read_excel(r"C:\Users\juiry\Documents\VKR\X_bp.xlsx", index_col=0)
df_nup = pd.read_excel(r"C:\Users\juiry\Documents\VKR\X_nup.xlsx", index_col=0)

df = df_bp.merge(df_nup, left_index=True,right_index=True, how='inner')

df


df.shape

(1023, 13)


df.isna().sum()

Соотношение матрица-наполнитель         0
Плотность, кг/м3                        0
модуль упругости, ГПа                   0
Количество отвердителя, м.%             0
Содержание эпоксидных групп,%_2         0
Температура вспышки, С_2                0
Поверхностная плотность, г/м2           0
Модуль упругости при растяжении, ГПа    0
Прочность при растяжении, МПа           0
Потребление смолы, г/м2                 0
Угол нашивки, град                      0
Шаг нашивки                             0
Плотность нашивки                       0
dtype: int64


sns.heatmap(df.isnull(), yticklabels=False, cbar=False, cmap='viridis');


le = LabelEncoder()
df['Угол нашивки, град'] = le.fit_transform(df['Угол нашивки, град'])
df


df.info();

<class 'pandas.core.frame.DataFrame'>
Float64Index: 1023 entries, 0.0 to 1022.0
Data columns (total 13 columns):
 #   Column                                Non-Null Count  Dtype  
---  ------                                --------------  -----  
 0   Соотношение матрица-наполнитель       1023 non-null   float64
 1   Плотность, кг/м3                      1023 non-null   float64
 2   модуль упругости, ГПа                 1023 non-null   float64
 3   Количество отвердителя, м.%           1023 non-null   float64
 4   Содержание эпоксидных групп,%_2       1023 non-null   float64
 5   Температура вспышки, С_2              1023 non-null   float64
 6   Поверхностная плотность, г/м2         1023 non-null   float64
 7   Модуль упругости при растяжении, ГПа  1023 non-null   float64
 8   Прочность при растяжении, МПа         1023 non-null   float64
 9   Потребление смолы, г/м2               1023 non-null   float64
 10  Угол нашивки, град                    1023 non-null   int64  
 11  Шаг нашивки                           1023 non-null   float64
 12  Плотность нашивки                     1023 non-null   float64
dtypes: float64(12), int64(1)
memory usage: 111.9 KB


df.describe()


scatter_matrix(df, alpha=0.2, figsize=(40, 40), diagonal='kde', color='black');


sns.set(rc = {'figure.figsize':(15, 8)})

# Визуализируем данные корреляции
sns.heatmap(df.corr(), annot=True);


df.hist(figsize=(20,15), color='black');


for col in df.columns:
    sns.catplot(x=col, data=df, kind='box', color='white', aspect=1*2)
    plt.title(col, fontsize=15)


num_rows = len(df.index)
low_information_cols = []

for col in df.columns:
    cnts = df[col].value_counts(dropna=False)
    top_pct = (cnts/num_rows).iloc[0]
    
    if top_pct > 0.95:
        low_information_cols.append(col)
        print('{0}: {1:5f}%'.format(col, top_pct*100))
        print(cnts)
        print()


for x in df.columns:
    q75,q25 = np.percentile(df.loc[:,x],[75,25])
    intr_qr = q75-q25
    
    max = q75+(1.5*intr_qr)
    min = q25-(1.5*intr_qr)
    
    df.loc[df[x] < min,x] = np.nan
    df.loc[df[x] > max,x] = np.nan
    
df.isnull().sum()

Соотношение матрица-наполнитель          6
Плотность, кг/м3                         9
модуль упругости, ГПа                    2
Количество отвердителя, м.%             14
Содержание эпоксидных групп,%_2          2
Температура вспышки, С_2                 8
Поверхностная плотность, г/м2            2
Модуль упругости при растяжении, ГПа     6
Прочность при растяжении, МПа           11
Потребление смолы, г/м2                  8
Угол нашивки, град                       0
Шаг нашивки                              4
Плотность нашивки                       21
dtype: int64

df


df_clean = df.dropna(axis=0)
df_clean


df_median = df.fillna(df.median())
df_median


scaler = MinMaxScaler()
df_transf = pd.DataFrame(scaler.fit_transform(df_clean))
df_norm = df_transf.set_axis(['Соотношение матрица-наполнитель', 'Плотность', 'Модуль упругости', 'Количество отвердителя', 'Содержание эпоксидных групп', 'Температура вспышки', 'Поверхностная плотность', 'Модуль упругости при растяжении', 'Прочность при растяжении', 'Потребление смолы', 'Угол нашивки', 'Шаг нашивки', 'Плотность нашивки'], axis=1, inplace=False)
df_norm.info();

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 936 entries, 0 to 935
Data columns (total 13 columns):
 #   Column                           Non-Null Count  Dtype  
---  ------                           --------------  -----  
 0   Соотношение матрица-наполнитель  936 non-null    float64
 1   Плотность                        936 non-null    float64
 2   Модуль упругости                 936 non-null    float64
 3   Количество отвердителя           936 non-null    float64
 4   Содержание эпоксидных групп      936 non-null    float64
 5   Температура вспышки              936 non-null    float64
 6   Поверхностная плотность          936 non-null    float64
 7   Модуль упругости при растяжении  936 non-null    float64
 8   Прочность при растяжении         936 non-null    float64
 9   Потребление смолы                936 non-null    float64
 10  Угол нашивки                     936 non-null    float64
 11  Шаг нашивки                      936 non-null    float64
 12  Плотность нашивки                936 non-null    float64
dtypes: float64(13)
memory usage: 95.2 KB


df_transf = pd.DataFrame(scaler.fit_transform(df_median))
df_norm_median = df_transf.set_axis(['Соотношение матрица-наполнитель', 'Плотность', 'Модуль упругости', 'Количество отвердителя', 'Содержание эпоксидных групп', 'Температура вспышки', 'Поверхностная плотность', 'Модуль упругости при растяжении', 'Прочность при растяжении', 'Потребление смолы', 'Угол нашивки', 'Шаг нашивки', 'Плотность нашивки'], axis=1, inplace=False)
df_norm_median


for col in df_norm.columns:
    sns.catplot(x=col, data=df_norm, kind='box', color='white', aspect=1*2)
    plt.title(col, fontsize=15)


df_norm.hist(figsize=(20,15), color='black');


X_1 = df_norm.drop(['Модуль упругости при растяжении'], axis=1)

Y_1 = df_norm['Модуль упругости при растяжении']

X_1.head()


X_2 = df_norm.drop(['Прочность при растяжении'], axis=1)

Y_2 = df_norm['Прочность при растяжении']

X_2.head()


X_3 = df_norm.drop(['Соотношение матрица-наполнитель'], axis=1)
X_3_median = df_norm_median.drop(['Соотношение матрица-наполнитель'], axis=1)

Y_3 = df_norm['Соотношение матрица-наполнитель']
Y_3_median = df_norm_median['Соотношение матрица-наполнитель']

X_3.head()


X_1_train, X_1_test, Y_1_train, Y_1_test = train_test_split(X_1, Y_1, test_size = 0.3, random_state=42)

X_2_train, X_2_test, Y_2_train, Y_2_test = train_test_split(X_2, Y_2, test_size = 0.3, random_state=42)

X_3_train, X_3_test, Y_3_train, Y_3_test = train_test_split(X_3, Y_3, test_size = 0.3, random_state=42)
X_3_train_median, X_3_test_median, Y_3_train_median, Y_3_test_median = train_test_split(X_3_median, Y_3_median, test_size = 0.3, random_state=42)


def mean_model(test_model):
    return [np.mean(test_model) for _ in range(len(test_model))]

Y_1_predict_mean = mean_model(Y_1_test)
Y_2_predict_mean = mean_model(Y_2_test)


slr = LinearRegression()

slr.fit(X_1_train, Y_1_train)
Y_1_train_predict = slr.predict(X_1_train)
Y_1_test_predict = slr.predict(X_1_test)

df_lin = pd.DataFrame({'Actual': Y_1_test, 'Predicted': Y_1_test_predict})
df_lin.head()


round(mean_squared_error(Y_1_test, Y_1_test_predict), 6)

0.034808


round(r2_score(Y_1_test, Y_1_test_predict), 6)

-0.005222


plt.axes(aspect='equal')
plt.scatter(Y_1_test, Y_1_test_predict, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


# график распределения ошибки

error = Y_1_test - Y_1_test_predict
plt.hist(error, bins = 25, color='black')
plt.xlabel('Prediction error')
plt.ylabel('Count');


round(mean_squared_error(Y_1_test, Y_1_predict_mean), 6)

0.034627


round(r2_score(Y_1_test, Y_1_predict_mean), 6)

0.0


plt.axes(aspect='equal')
plt.scatter(Y_1_test, Y_1_predict_mean, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


dtr = DecisionTreeRegressor(random_state=1)

parameters_dtr = {'max_depth': range (1,7),    
              'min_samples_leaf': range (1,8),
              'min_samples_split': range (2,10,2) }

grid_dtr = GridSearchCV(dtr, parameters_dtr, cv=10)
grid_dtr.fit(X_1_train, Y_1_train)

grid_dtr.best_params_

{'max_depth': 1, 'min_samples_leaf': 6, 'min_samples_split': 2}


decision_tree_model = DecisionTreeRegressor(max_depth = 1, min_samples_leaf = 6, min_samples_split = 2, random_state=1)
decision_tree_model.fit(X_1_train, Y_1_train)
Y_predict_tree = decision_tree_model.predict(X_1_test)

round(mean_squared_error(Y_1_test, Y_predict_tree), 6)

0.034693


round(r2_score(Y_1_test, Y_predict_tree), 6)

-0.001902


plt.axes(aspect='equal')
plt.scatter(Y_1_test, Y_predict_tree, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


rfr = RandomForestRegressor()

parametrs_rfr = { 'n_estimators': range (10, 51, 5),
              'max_depth': range (1,7),
              'min_samples_leaf': range (1,8)}

grid_rfr = GridSearchCV(rfr, parametrs_rfr, cv=10)
grid_rfr.fit(X_1_train, Y_1_train)

grid_rfr.best_params_

{'max_depth': 1, 'min_samples_leaf': 2, 'n_estimators': 10}


rand_forest_model = RandomForestRegressor(n_estimators = 15, max_depth = 1, min_samples_leaf = 7, min_samples_split=4)
rand_forest_model.fit(X_1_train, Y_1_train)
Y_predict_forest = rand_forest_model.predict(X_1_test)

round(mean_squared_error(Y_1_test, Y_predict_forest), 6)

0.034793


round(r2_score(Y_1_test, Y_predict_forest), 6)

-0.004805


plt.axes(aspect='equal')
plt.scatter(Y_1_test, Y_predict_forest, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


gbr = GradientBoostingRegressor()

parameters_gbr = {'n_estimators': range (30, 51, 10),
              'max_depth': range (1,2),
              'min_samples_leaf': range (1,5)}

grid_gbr = GridSearchCV(gbr, parameters_gbr, cv=10)
grid_gbr.fit(X_1_train, Y_1_train)

grid_gbr.best_params_

{'max_depth': 1, 'min_samples_leaf': 1, 'n_estimators': 30}


gradient_boosting_model = GradientBoostingRegressor(n_estimators = 50, max_depth = 1, min_samples_leaf = 3)
gradient_boosting_model.fit(X_1_train, Y_1_train)
Y_predict_boosting = gradient_boosting_model.predict(X_1_test)

round(mean_squared_error(Y_1_test, Y_predict_boosting), 6)

0.034329


round(r2_score(Y_1_test, Y_predict_boosting), 6)

0.00862


plt.axes(aspect='equal')
plt.scatter(Y_1_test, Y_predict_boosting, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


mse_compare = pd.DataFrame({'Model': ['Linear Regressor', 'Decision Tree Regressor', 'Random Forest Regressor',
                                     'Gradient Boosting Regressor', 'Mean Function'], 
                            'MSE': [0.034808, 0.034693, 0.034701, 0.034329, 0.034627]}).sort_values('MSE', axis=0, ascending=True, inplace=False)
mse_compare


slr = LinearRegression()

slr.fit(X_2_train, Y_2_train)
Y_2_train_predict = slr.predict(X_2_train)
Y_2_test_predict = slr.predict(X_2_test)

df_lin = pd.DataFrame({'Actual': Y_2_test, 'Predicted': Y_2_test_predict})
df_lin.head()


round(mean_squared_error(Y_2_test, Y_2_test_predict), 6)

0.035449


round(r2_score(Y_2_test, Y_2_test_predict), 6)

-0.046999


plt.axes(aspect='equal')
plt.scatter(Y_2_test, Y_2_test_predict, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


# график распределения ошибки

error = Y_2_test - Y_2_test_predict
plt.hist(error, bins = 25, color='black')
plt.xlabel('Prediction error')
plt.ylabel('Count');


round(mean_squared_error(Y_2_test, Y_2_predict_mean), 6)

0.033857


round(r2_score(Y_2_test, Y_2_predict_mean), 6)

-0.0


plt.axes(aspect='equal')
plt.scatter(Y_2_test, Y_2_predict_mean, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


dtr = DecisionTreeRegressor(random_state=1)

parameters_dtr = {'max_depth': range (1,7),    
              'min_samples_leaf': range (1,8),
              'min_samples_split': range (2,10,2) }

grid_dtr = GridSearchCV(dtr, parameters_dtr, cv=10)
grid_dtr.fit(X_2_train, Y_2_train)

grid_dtr.best_params_

{'max_depth': 1, 'min_samples_leaf': 1, 'min_samples_split': 2}


decision_tree_model = DecisionTreeRegressor(max_depth = 1, min_samples_leaf = 1, min_samples_split = 2, random_state=1)
decision_tree_model.fit(X_2_train, Y_2_train)
Y_predict_tree = decision_tree_model.predict(X_2_test)

round(mean_squared_error(Y_2_test, Y_predict_tree), 6)

0.035352


round(r2_score(Y_2_test, Y_predict_tree), 6)

-0.044133


plt.axes(aspect='equal')
plt.scatter(Y_2_test, Y_predict_tree, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


rfr = RandomForestRegressor()

parametrs_rfr = { 'n_estimators': range (10, 31, 10),
              'max_depth': range (1,7),
              'min_samples_leaf': range (1,8)}

grid_rfr = GridSearchCV(rfr, parametrs_rfr, cv=10)
grid_rfr.fit(X_2_train, Y_2_train)

grid_rfr.best_params_

{'max_depth': 1, 'min_samples_leaf': 5, 'n_estimators': 10}


rand_forest_model = RandomForestRegressor(n_estimators = 10, max_depth = 1, min_samples_leaf = 1)
rand_forest_model.fit(X_2_train, Y_2_train)
Y_predict_forest = rand_forest_model.predict(X_2_test)

round(mean_squared_error(Y_2_test, Y_predict_forest), 6)

0.035063


round(r2_score(Y_2_test, Y_predict_forest), 6)

-0.035611


plt.axes(aspect='equal')
plt.scatter(Y_2_test, Y_predict_forest, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


gbr = GradientBoostingRegressor()

parameters_gbr = {'n_estimators': range (30, 51, 10),
              'max_depth': range (1,2),
              'min_samples_leaf': range (1,5)}

grid_gbr = GridSearchCV(gbr, parameters_gbr, cv=10)
grid_gbr.fit(X_2_train, Y_2_train)

grid_gbr.best_params_

{'max_depth': 1, 'min_samples_leaf': 3, 'n_estimators': 30}


gradient_boosting_model = GradientBoostingRegressor(n_estimators = 30, max_depth = 1, min_samples_leaf = 3)
gradient_boosting_model.fit(X_1_train, Y_1_train)
Y_predict_boosting = gradient_boosting_model.predict(X_1_test)

round(mean_squared_error(Y_1_test, Y_predict_boosting), 6)

0.034368


round(r2_score(Y_2_test, Y_predict_boosting), 6)

-0.033789


plt.axes(aspect='equal')
plt.scatter(Y_2_test, Y_predict_boosting, color='black')
plt.xlabel('Actual')
plt.ylabel('Predicted')
plt.plot([0, 1],[0, 1]);


mse_compare = pd.DataFrame({'Model': ['Linear Regressor', 'Decision Tree Regressor', 'Random Forest Regressor',
                                     'Gradient Boosting Regressor', 'Mean Function'], 
                            'MSE': [0.035449, 0.035352, 0.035164, 0.034368, 0.033857]}).sort_values('MSE', axis=0, ascending=True, inplace=False)
mse_compare


n_model = keras.Sequential([
    layers.Dense(12, activation = 'relu', input_dim=12),
    layers.Dense(128, activation = 'relu'),
    layers.Dense(64, activation = 'relu'),
    layers.Dense(64, activation = 'relu'),
    layers.Dense(1),
])

# изменение количества слоев и нейронов выше данных значение практически не показало изменения результата


n_model.compile(optimizer='SGD',
             loss='mse',
             metrics='mse')

# оптимайзеры Adam, RMSProp показали результаты хуже и были склонны к переобучению (например, Adam увеличивал MSE на тестовой
# выборке уже начиная с 15 эпохи)


n_model.summary()

Model: "sequential"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 dense (Dense)               (None, 12)                156       
                                                                 
 dense_1 (Dense)             (None, 128)               1664      
                                                                 
 dense_2 (Dense)             (None, 64)                8256      
                                                                 
 dense_3 (Dense)             (None, 64)                4160      
                                                                 
 dense_4 (Dense)             (None, 1)                 65        
                                                                 
=================================================================
Total params: 14,301
Trainable params: 14,301
Non-trainable params: 0
_________________________________________________________________


history = n_model.fit(X_3_train, Y_3_train, epochs=100, validation_split = 0.3)

# результаты остаются стабильными и при прохождении 500 эпох

Epoch 1/100
15/15 [==============================] - 1s 14ms/step - loss: 0.1779 - mse: 0.1779 - val_loss: 0.0551 - val_mse: 0.0551
Epoch 2/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0463 - mse: 0.0463 - val_loss: 0.0379 - val_mse: 0.0379
Epoch 3/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0384 - mse: 0.0384 - val_loss: 0.0376 - val_mse: 0.0376
Epoch 4/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0380 - mse: 0.0380 - val_loss: 0.0376 - val_mse: 0.0376
Epoch 5/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0377 - mse: 0.0377 - val_loss: 0.0374 - val_mse: 0.0374
Epoch 6/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0377 - mse: 0.0377 - val_loss: 0.0377 - val_mse: 0.0377
Epoch 7/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0375 - mse: 0.0375 - val_loss: 0.0375 - val_mse: 0.0375
Epoch 8/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0373 - mse: 0.0373 - val_loss: 0.0371 - val_mse: 0.0371
Epoch 9/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0373 - mse: 0.0373 - val_loss: 0.0375 - val_mse: 0.0375
Epoch 10/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0371 - mse: 0.0371 - val_loss: 0.0372 - val_mse: 0.0372
Epoch 11/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0370 - mse: 0.0370 - val_loss: 0.0368 - val_mse: 0.0368
Epoch 12/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0370 - mse: 0.0370 - val_loss: 0.0369 - val_mse: 0.0369
Epoch 13/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0369 - mse: 0.0369 - val_loss: 0.0370 - val_mse: 0.0370
Epoch 14/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0367 - mse: 0.0367 - val_loss: 0.0370 - val_mse: 0.0370
Epoch 15/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0368 - mse: 0.0368 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 16/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0367 - mse: 0.0367 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 17/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0366 - mse: 0.0366 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 18/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0366 - mse: 0.0366 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 19/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0365 - mse: 0.0365 - val_loss: 0.0372 - val_mse: 0.0372
Epoch 20/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0365 - mse: 0.0365 - val_loss: 0.0367 - val_mse: 0.0367
Epoch 21/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0364 - mse: 0.0364 - val_loss: 0.0364 - val_mse: 0.0364
Epoch 22/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0364 - mse: 0.0364 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 23/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0364 - mse: 0.0364 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 24/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0363 - mse: 0.0363 - val_loss: 0.0364 - val_mse: 0.0364
Epoch 25/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0363 - mse: 0.0363 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 26/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0362 - mse: 0.0362 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 27/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0362 - mse: 0.0362 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 28/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0362 - mse: 0.0362 - val_loss: 0.0368 - val_mse: 0.0368
Epoch 29/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0362 - mse: 0.0362 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 30/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0361 - mse: 0.0361 - val_loss: 0.0363 - val_mse: 0.0363
Epoch 31/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0360 - mse: 0.0360 - val_loss: 0.0364 - val_mse: 0.0364
Epoch 32/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0361 - mse: 0.0361 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 33/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0360 - mse: 0.0360 - val_loss: 0.0365 - val_mse: 0.0365
Epoch 34/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0361 - mse: 0.0361 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 35/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0360 - mse: 0.0360 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 36/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0360 - mse: 0.0360 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 37/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0360 - mse: 0.0360 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 38/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0360 - mse: 0.0360 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 39/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0359 - mse: 0.0359 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 40/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0359 - mse: 0.0359 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 41/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0359 - mse: 0.0359 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 42/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0359 - mse: 0.0359 - val_loss: 0.0363 - val_mse: 0.0363
Epoch 43/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0357 - mse: 0.0357 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 44/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0358 - mse: 0.0358 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 45/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0357 - mse: 0.0357 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 46/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0358 - mse: 0.0358 - val_loss: 0.0364 - val_mse: 0.0364
Epoch 47/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0357 - mse: 0.0357 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 48/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0356 - mse: 0.0356 - val_loss: 0.0357 - val_mse: 0.0357
Epoch 49/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0357 - mse: 0.0357 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 50/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0357 - mse: 0.0357 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 51/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0356 - mse: 0.0356 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 52/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0357 - mse: 0.0357 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 53/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0356 - mse: 0.0356 - val_loss: 0.0363 - val_mse: 0.0363
Epoch 54/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0356 - mse: 0.0356 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 55/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0355 - mse: 0.0355 - val_loss: 0.0357 - val_mse: 0.0357
Epoch 56/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0355 - mse: 0.0355 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 57/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 58/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0355 - mse: 0.0355 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 59/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 60/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0355 - mse: 0.0355 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 61/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 62/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 63/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 64/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0355 - mse: 0.0355 - val_loss: 0.0367 - val_mse: 0.0367
Epoch 65/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0355 - mse: 0.0355 - val_loss: 0.0364 - val_mse: 0.0364
Epoch 66/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 67/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0354 - mse: 0.0354 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 68/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 69/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 70/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0356 - val_mse: 0.0356
Epoch 71/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 72/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 73/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0366 - val_mse: 0.0366
Epoch 74/100
15/15 [==============================] - 0s 4ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 75/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0362 - val_mse: 0.0362
Epoch 76/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 77/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0361 - val_mse: 0.0361
Epoch 78/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 79/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0356 - val_mse: 0.0356
Epoch 80/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0353 - mse: 0.0353 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 81/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0356 - val_mse: 0.0356
Epoch 82/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 83/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 84/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 85/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0352 - mse: 0.0352 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 86/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0364 - val_mse: 0.0364
Epoch 87/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0356 - val_mse: 0.0356
Epoch 88/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 89/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 90/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 91/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0357 - val_mse: 0.0357
Epoch 92/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0357 - val_mse: 0.0357
Epoch 93/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0360 - val_mse: 0.0360
Epoch 94/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0350 - mse: 0.0350 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 95/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 96/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0356 - val_mse: 0.0356
Epoch 97/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0351 - mse: 0.0351 - val_loss: 0.0358 - val_mse: 0.0358
Epoch 98/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0350 - mse: 0.0350 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 99/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0350 - mse: 0.0350 - val_loss: 0.0359 - val_mse: 0.0359
Epoch 100/100
15/15 [==============================] - 0s 3ms/step - loss: 0.0350 - mse: 0.0350 - val_loss: 0.0360 - val_mse: 0.0360


n_model.evaluate(X_3_test, Y_3_test, verbose=1)

9/9 [==============================] - 0s 1ms/step - loss: 0.0338 - mse: 0.0338

[0.03383169323205948, 0.03383169323205948]


n_model.save("neural_net")

INFO:tensorflow:Assets written to: neural_net\assets


predict_model =  n_model.predict(X_3_test)
predict_values = np.ravel(predict_model) 
predict_values

array([0.508, 0.509, 0.523, 0.504, 0.469, 0.506, 0.485, 0.519, 0.491,
       0.566, 0.508, 0.579, 0.54 , 0.498, 0.487, 0.503, 0.485, 0.488,
       0.512, 0.506, 0.488, 0.513, 0.501, 0.505, 0.548, 0.531, 0.498,
       0.487, 0.509, 0.523, 0.495, 0.502, 0.467, 0.496, 0.498, 0.479,
       0.497, 0.514, 0.501, 0.486, 0.499, 0.607, 0.499, 0.48 , 0.524,
       0.508, 0.504, 0.577, 0.504, 0.499, 0.528, 0.532, 0.513, 0.491,
       0.5  , 0.476, 0.529, 0.538, 0.496, 0.523, 0.478, 0.49 , 0.498,
       0.481, 0.501, 0.505, 0.482, 0.489, 0.533, 0.503, 0.505, 0.516,
       0.544, 0.516, 0.509, 0.534, 0.51 , 0.55 , 0.533, 0.515, 0.57 ,
       0.472, 0.488, 0.486, 0.539, 0.556, 0.5  , 0.503, 0.517, 0.494,
       0.547, 0.576, 0.494, 0.537, 0.54 , 0.528, 0.552, 0.512, 0.484,
       0.537, 0.511, 0.513, 0.497, 0.505, 0.508, 0.509, 0.495, 0.497,
       0.473, 0.502, 0.498, 0.563, 0.555, 0.584, 0.516, 0.529, 0.513,
       0.512, 0.497, 0.483, 0.492, 0.509, 0.498, 0.51 , 0.477, 0.484,
       0.497, 0.549, 0.532, 0.481, 0.492, 0.553, 0.517, 0.508, 0.497,
       0.497, 0.483, 0.527, 0.536, 0.481, 0.487, 0.493, 0.5  , 0.518,
       0.523, 0.477, 0.511, 0.502, 0.498, 0.457, 0.506, 0.511, 0.507,
       0.479, 0.528, 0.505, 0.552, 0.496, 0.549, 0.5  , 0.492, 0.498,
       0.501, 0.466, 0.512, 0.532, 0.501, 0.539, 0.472, 0.52 , 0.512,
       0.478, 0.514, 0.501, 0.496, 0.481, 0.515, 0.485, 0.581, 0.522,
       0.493, 0.556, 0.513, 0.519, 0.532, 0.587, 0.525, 0.553, 0.486,
       0.548, 0.559, 0.502, 0.588, 0.487, 0.558, 0.5  , 0.498, 0.49 ,
       0.497, 0.499, 0.516, 0.477, 0.485, 0.529, 0.528, 0.54 , 0.533,
       0.5  , 0.541, 0.511, 0.469, 0.528, 0.49 , 0.504, 0.504, 0.535,
       0.519, 0.575, 0.555, 0.491, 0.585, 0.498, 0.505, 0.513, 0.495,
       0.489, 0.465, 0.477, 0.507, 0.526, 0.508, 0.524, 0.542, 0.488,
       0.493, 0.514, 0.506, 0.504, 0.554, 0.488, 0.519, 0.518, 0.539,
       0.509, 0.518, 0.477, 0.495, 0.531, 0.517, 0.518, 0.524, 0.517,
       0.5  , 0.537, 0.513, 0.519, 0.496, 0.547, 0.518, 0.493, 0.497,
       0.48 , 0.532, 0.513, 0.497, 0.492, 0.508, 0.571, 0.574, 0.523,
       0.502, 0.534, 0.503, 0.516, 0.532, 0.48 , 0.47 , 0.482, 0.535,
       0.512, 0.473], dtype=float32)


Y_3_test_values = np.ravel(Y_3_test.copy())
Y_3_test_values

array([0.488, 0.292, 0.246, 0.851, 0.611, 0.448, 0.205, 0.63 , 0.425,
       0.384, 0.315, 0.399, 0.6  , 0.411, 0.616, 0.535, 0.39 , 0.34 ,
       0.209, 0.471, 0.346, 0.742, 0.662, 0.511, 0.419, 0.865, 0.27 ,
       0.578, 0.789, 0.532, 0.433, 0.418, 0.136, 0.313, 0.51 , 0.619,
       0.705, 0.199, 0.672, 0.715, 0.448, 0.417, 0.519, 0.245, 0.336,
       0.436, 0.606, 0.522, 0.801, 0.443, 0.712, 0.449, 0.59 , 0.405,
       0.275, 0.58 , 0.996, 0.393, 0.595, 0.817, 0.088, 0.524, 0.825,
       0.588, 0.198, 0.497, 0.636, 0.252, 0.095, 0.728, 0.404, 0.651,
       0.554, 0.492, 0.479, 0.815, 0.52 , 0.333, 0.477, 0.306, 0.428,
       0.319, 0.671, 0.462, 0.292, 0.491, 0.209, 0.36 , 0.773, 0.453,
       0.321, 0.589, 0.451, 0.336, 0.835, 0.243, 0.267, 0.263, 0.551,
       0.811, 0.594, 0.281, 0.305, 0.51 , 0.442, 0.804, 0.513, 0.464,
       0.737, 0.663, 0.556, 0.631, 0.513, 0.48 , 0.674, 0.34 , 0.346,
       0.448, 0.375, 0.515, 0.362, 0.606, 0.75 , 0.597, 0.376, 0.268,
       0.444, 0.347, 0.629, 0.544, 0.537, 0.331, 0.394, 0.495, 0.701,
       0.572, 0.308, 0.432, 0.796, 0.494, 0.859, 0.13 , 0.422, 0.349,
       0.467, 0.589, 0.704, 0.667, 0.479, 0.589, 0.59 , 0.463, 0.104,
       0.268, 0.597, 0.159, 0.327, 0.576, 0.647, 0.512, 0.538, 0.652,
       0.637, 0.45 , 0.136, 0.464, 0.749, 0.498, 0.394, 0.497, 0.672,
       0.279, 0.366, 0.525, 0.38 , 0.35 , 0.505, 0.386, 0.495, 0.582,
       0.135, 0.327, 0.25 , 0.735, 0.467, 0.41 , 0.473, 0.691, 0.423,
       0.539, 0.344, 0.65 , 0.458, 0.51 , 0.957, 0.63 , 0.543, 0.723,
       0.644, 0.339, 0.146, 0.454, 0.335, 0.22 , 0.426, 0.55 , 0.469,
       0.319, 0.246, 0.464, 0.487, 0.849, 0.573, 0.447, 0.472, 0.881,
       0.291, 0.494, 0.893, 0.491, 0.415, 0.57 , 0.398, 0.381, 0.069,
       0.526, 0.213, 0.499, 0.582, 0.486, 0.555, 0.549, 0.409, 0.78 ,
       0.839, 0.404, 0.621, 0.221, 0.273, 0.489, 0.394, 0.423, 0.572,
       0.715, 0.872, 0.059, 0.777, 0.208, 0.626, 0.641, 0.362, 0.54 ,
       0.68 , 0.644, 0.316, 0.397, 0.565, 0.554, 0.51 , 0.781, 0.617,
       0.478, 0.42 , 0.685, 0.774, 0.15 , 0.462, 0.7  , 0.74 , 0.508,
       0.39 , 0.414, 0.722, 0.239, 0.82 , 0.381, 0.528, 0.441, 0.835,
       0.535, 0.569])


feature_results = pd.DataFrame({'Данные': Y_3_test_values,'Прогноз': predict_values})
sns.set_style('darkgrid')
plt.title('Рассеяние тестовых и прогнозных значений',size=16)
plt.xlabel('Данные',size=12)
plt.ylabel('Прогноз',size=12)
sns.scatterplot(x='Данные', y='Прогноз', data=feature_results, color="black")
plt.show()
feature_results


history_median = n_model.fit(X_3_train_median, Y_3_train_median, epochs=100, validation_split = 0.3)

Epoch 1/100
16/16 [==============================] - 0s 6ms/step - loss: 0.0347 - mse: 0.0347 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 2/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 3/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 4/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 5/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 6/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 7/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 8/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 9/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 10/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 11/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 12/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 13/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 14/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 15/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 16/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0346 - mse: 0.0346 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 17/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 18/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 19/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 20/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 21/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0345 - mse: 0.0345 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 22/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 23/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 24/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0323 - val_mse: 0.0323
Epoch 25/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 26/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 27/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 28/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 29/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 30/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 31/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 32/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 33/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 34/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 35/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 36/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 37/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 38/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 39/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 40/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 41/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 42/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 43/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 44/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 45/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 46/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 47/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 48/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 49/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 50/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 51/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 52/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 53/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0344 - mse: 0.0344 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 54/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 55/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 56/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 57/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 58/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 59/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 60/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 61/100
16/16 [==============================] - 0s 4ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0327 - val_mse: 0.0327
Epoch 62/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 63/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 64/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 65/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 66/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 67/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 68/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 69/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 70/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0327 - val_mse: 0.0327
Epoch 71/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 72/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 73/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 74/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 75/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 76/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 77/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 78/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 79/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 80/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 81/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0324 - val_mse: 0.0324
Epoch 82/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 83/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 84/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 85/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 86/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 87/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0343 - mse: 0.0343 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 88/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 89/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0341 - mse: 0.0341 - val_loss: 0.0329 - val_mse: 0.0329
Epoch 90/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0327 - val_mse: 0.0327
Epoch 91/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 92/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 93/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 94/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 95/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 96/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 97/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0325 - val_mse: 0.0325
Epoch 98/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0326 - val_mse: 0.0326
Epoch 99/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0327 - val_mse: 0.0327
Epoch 100/100
16/16 [==============================] - 0s 3ms/step - loss: 0.0342 - mse: 0.0342 - val_loss: 0.0328 - val_mse: 0.0328


n_model.evaluate(X_3_test_median, Y_3_test_median, verbose=1)

10/10 [==============================] - 0s 1ms/step - loss: 0.0366 - mse: 0.0366

[0.036648865789175034, 0.036648865789175034]

	Соотношение матрица-наполнитель	Плотность, кг/м3	модуль упругости, ГПа	Количество отвердителя, м.%	Содержание эпоксидных групп,%_2	Температура вспышки, С_2	Поверхностная плотность, г/м2	Модуль упругости при растяжении, ГПа	Прочность при растяжении, МПа	Потребление смолы, г/м2	Угол нашивки, град	Шаг нашивки	Плотность нашивки
0.0	1.857143	2030.000000	738.736842	30.000000	22.267857	100.000000	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	57.000000
1.0	1.857143	2030.000000	738.736842	50.000000	23.750000	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	60.000000
2.0	1.857143	2030.000000	738.736842	49.900000	33.000000	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	70.000000
3.0	1.857143	2030.000000	738.736842	129.000000	21.250000	300.000000	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	47.000000
4.0	2.771331	2030.000000	753.000000	111.860000	22.267857	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	57.000000
...	...	...	...	...	...	...	...	...	...	...	...	...	...
1018.0	2.271346	1952.087902	912.855545	86.992183	20.123249	324.774576	209.198700	73.090961	2387.292495	125.007669	90.0	9.076380	47.019770
1019.0	3.444022	2050.089171	444.732634	145.981978	19.599769	254.215401	350.660830	72.920827	2360.392784	117.730099	90.0	10.565614	53.750790
1020.0	3.280604	1972.372865	416.836524	110.533477	23.957502	248.423047	740.142791	74.734344	2662.906040	236.606764	90.0	4.161154	67.629684
1021.0	3.705351	2066.799773	741.475517	141.397963	19.246945	275.779840	641.468152	74.042708	2071.715856	197.126067	90.0	6.313201	58.261074
1022.0	3.808020	1890.413468	417.316232	129.183416	27.474763	300.952708	758.747882	74.309704	2856.328932	194.754342	90.0	6.078902	77.434468

	Соотношение матрица-наполнитель	Плотность, кг/м3	модуль упругости, ГПа	Количество отвердителя, м.%	Содержание эпоксидных групп,%_2	Температура вспышки, С_2	Поверхностная плотность, г/м2	Модуль упругости при растяжении, ГПа	Прочность при растяжении, МПа	Потребление смолы, г/м2	Угол нашивки, град	Шаг нашивки	Плотность нашивки
0.0	1.857143	2030.000000	738.736842	30.000000	22.267857	100.000000	210.000000	70.000000	3000.000000	220.000000	0	4.000000	57.000000
1.0	1.857143	2030.000000	738.736842	50.000000	23.750000	284.615385	210.000000	70.000000	3000.000000	220.000000	0	4.000000	60.000000
2.0	1.857143	2030.000000	738.736842	49.900000	33.000000	284.615385	210.000000	70.000000	3000.000000	220.000000	0	4.000000	70.000000
3.0	1.857143	2030.000000	738.736842	129.000000	21.250000	300.000000	210.000000	70.000000	3000.000000	220.000000	0	5.000000	47.000000
4.0	2.771331	2030.000000	753.000000	111.860000	22.267857	284.615385	210.000000	70.000000	3000.000000	220.000000	0	5.000000	57.000000
...	...	...	...	...	...	...	...	...	...	...	...	...	...
1018.0	2.271346	1952.087902	912.855545	86.992183	20.123249	324.774576	209.198700	73.090961	2387.292495	125.007669	1	9.076380	47.019770
1019.0	3.444022	2050.089171	444.732634	145.981978	19.599769	254.215401	350.660830	72.920827	2360.392784	117.730099	1	10.565614	53.750790
1020.0	3.280604	1972.372865	416.836524	110.533477	23.957502	248.423047	740.142791	74.734344	2662.906040	236.606764	1	4.161154	67.629684
1021.0	3.705351	2066.799773	741.475517	141.397963	19.246945	275.779840	641.468152	74.042708	2071.715856	197.126067	1	6.313201	58.261074
1022.0	3.808020	1890.413468	417.316232	129.183416	27.474763	300.952708	758.747882	74.309704	2856.328932	194.754342	1	6.078902	77.434468

	Соотношение матрица-наполнитель	Плотность, кг/м3	модуль упругости, ГПа	Количество отвердителя, м.%	Содержание эпоксидных групп,%_2	Температура вспышки, С_2	Поверхностная плотность, г/м2	Модуль упругости при растяжении, ГПа	Прочность при растяжении, МПа	Потребление смолы, г/м2	Угол нашивки, град	Шаг нашивки	Плотность нашивки
count	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000	1023.000000
mean	2.930366	1975.734888	739.923233	110.570769	22.244390	285.882151	482.731833	73.328571	2466.922843	218.423144	0.491691	6.899222	57.153929
std	0.913222	73.729231	330.231581	28.295911	2.406301	40.943260	281.314690	3.118983	485.628006	59.735931	0.500175	2.563467	12.350969
min	0.389403	1731.764635	2.436909	17.740275	14.254985	100.000000	0.603740	64.054061	1036.856605	33.803026	0.000000	0.000000	0.000000
25%	2.317887	1924.155467	500.047452	92.443497	20.608034	259.066528	266.816645	71.245018	2135.850448	179.627520	0.000000	5.080033	49.799212
50%	2.906878	1977.621657	739.664328	110.564840	22.230744	285.896812	451.864365	73.268805	2459.524526	219.198882	0.000000	6.916144	57.341920
75%	3.552660	2021.374375	961.812526	129.730366	23.961934	313.002106	693.225017	75.356612	2767.193119	257.481724	1.000000	8.586293	64.944961
max	5.591742	2207.773481	1911.536477	198.953207	33.000000	413.273418	1399.542362	82.682051	3848.436732	414.590628	1.000000	14.440522	103.988901

	Соотношение матрица-наполнитель	Плотность, кг/м3	модуль упругости, ГПа	Количество отвердителя, м.%	Содержание эпоксидных групп,%_2	Температура вспышки, С_2	Поверхностная плотность, г/м2	Модуль упругости при растяжении, ГПа	Прочность при растяжении, МПа	Потребление смолы, г/м2	Угол нашивки, град	Шаг нашивки	Плотность нашивки
0.0	1.857143	2030.000000	738.736842	NaN	22.267857	NaN	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	57.000000
1.0	1.857143	2030.000000	738.736842	50.000000	23.750000	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	60.000000
2.0	1.857143	2030.000000	738.736842	49.900000	NaN	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	70.000000
3.0	1.857143	2030.000000	738.736842	129.000000	21.250000	300.000000	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	47.000000
4.0	2.771331	2030.000000	753.000000	111.860000	22.267857	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	57.000000
...	...	...	...	...	...	...	...	...	...	...	...	...	...
1018.0	2.271346	1952.087902	912.855545	86.992183	20.123249	324.774576	209.198700	73.090961	2387.292495	125.007669	1.0	9.076380	47.019770
1019.0	3.444022	2050.089171	444.732634	145.981978	19.599769	254.215401	350.660830	72.920827	2360.392784	117.730099	1.0	10.565614	53.750790
1020.0	3.280604	1972.372865	416.836524	110.533477	23.957502	248.423047	740.142791	74.734344	2662.906040	236.606764	1.0	4.161154	67.629684
1021.0	3.705351	2066.799773	741.475517	141.397963	19.246945	275.779840	641.468152	74.042708	2071.715856	197.126067	1.0	6.313201	58.261074
1022.0	3.808020	1890.413468	417.316232	129.183416	27.474763	300.952708	758.747882	74.309704	2856.328932	194.754342	1.0	6.078902	77.434468

	Соотношение матрица-наполнитель	Плотность, кг/м3	модуль упругости, ГПа	Количество отвердителя, м.%	Содержание эпоксидных групп,%_2	Температура вспышки, С_2	Поверхностная плотность, г/м2	Модуль упругости при растяжении, ГПа	Прочность при растяжении, МПа	Потребление смолы, г/м2	Угол нашивки, град	Шаг нашивки	Плотность нашивки
1.0	1.857143	2030.000000	738.736842	50.000000	23.750000	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	4.000000	60.000000
3.0	1.857143	2030.000000	738.736842	129.000000	21.250000	300.000000	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	47.000000
4.0	2.771331	2030.000000	753.000000	111.860000	22.267857	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	57.000000
5.0	2.767918	2000.000000	748.000000	111.860000	22.267857	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	60.000000
6.0	2.569620	1910.000000	807.000000	111.860000	22.267857	284.615385	210.000000	70.000000	3000.000000	220.000000	0.0	5.000000	70.000000
...	...	...	...	...	...	...	...	...	...	...	...	...	...
1018.0	2.271346	1952.087902	912.855545	86.992183	20.123249	324.774576	209.198700	73.090961	2387.292495	125.007669	1.0	9.076380	47.019770
1019.0	3.444022	2050.089171	444.732634	145.981978	19.599769	254.215401	350.660830	72.920827	2360.392784	117.730099	1.0	10.565614	53.750790
1020.0	3.280604	1972.372865	416.836524	110.533477	23.957502	248.423047	740.142791	74.734344	2662.906040	236.606764	1.0	4.161154	67.629684
1021.0	3.705351	2066.799773	741.475517	141.397963	19.246945	275.779840	641.468152	74.042708	2071.715856	197.126067	1.0	6.313201	58.261074
1022.0	3.808020	1890.413468	417.316232	129.183416	27.474763	300.952708	758.747882	74.309704	2856.328932	194.754342	1.0	6.078902	77.434468

Постановка задачи¶

Подключение библиотек¶

Загрузка данных¶

Разведочный анализ данных¶

Проверка наличия пропусков в данных¶

Кодируем угол нашивки в LabelEncoder¶

Квартили, среднее, медианное значение для параметров¶

Попарные графики рассеяния точек¶

Проверка корреляции признаков¶

Гистограммы:¶

Диаграмма "ящик с усами"¶

Предобработка данных¶

Исключение малоинформативных признаков¶

Исключение выбросов на основе межквартильного расстояния¶

Нормализация данных¶

Повторяем диаграмму ящик с усами¶

Повторяем график распределения¶

Подготовка моделей¶

Создание¶

Разделение на тренировочную и тестовую части¶

Создание модели, выдающей среднее значение для сравнения¶

Обучение моделей¶

Прогноз модуля упругости при растяжении¶

Прогноз прочности при растяжении¶

Прогноз соотношения матрица-наполнитель¶

	Соотношение матрица-наполнитель	Плотность	Модуль упругости	Количество отвердителя	Содержание эпоксидных групп	Температура вспышки	Поверхностная плотность	Модуль упругости при растяжении	Прочность при растяжении	Потребление смолы	Угол нашивки	Шаг нашивки	Плотность нашивки
0	0.274768	0.651097	0.447061	0.502210	0.495653	0.515364	0.162230	0.280303	0.712590	0.529221	0.0	0.291282	0.506083
1	0.274768	0.651097	0.447061	0.079153	0.607435	0.509164	0.162230	0.280303	0.712590	0.529221	0.0	0.291282	0.557156
2	0.274768	0.651097	0.447061	0.078454	0.492854	0.509164	0.162230	0.280303	0.712590	0.529221	0.0	0.291282	0.727399
3	0.274768	0.651097	0.447061	0.630983	0.418887	0.583596	0.162230	0.280303	0.712590	0.529221	0.0	0.364102	0.335840
4	0.466552	0.651097	0.455721	0.511257	0.495653	0.509164	0.162230	0.280303	0.712590	0.529221	0.0	0.364102	0.506083
...	...	...	...	...	...	...	...	...	...	...	...	...	...
1018	0.361662	0.444480	0.552781	0.337550	0.333908	0.703458	0.161609	0.475147	0.463043	0.207613	1.0	0.660946	0.336177
1019	0.607674	0.704373	0.268550	0.749605	0.294428	0.362087	0.271207	0.464422	0.452087	0.182974	1.0	0.769393	0.450768
1020	0.573391	0.498274	0.251612	0.501991	0.623085	0.334063	0.572959	0.578740	0.575296	0.585446	1.0	0.303017	0.687046
1021	0.662497	0.748688	0.448724	0.717585	0.267818	0.466417	0.496511	0.535142	0.334513	0.451779	1.0	0.459730	0.527552
1022	0.684036	0.280923	0.251903	0.632264	0.888354	0.588206	0.587373	0.551972	0.654075	0.443749	1.0	0.442668	0.853966

	Actual	Predicted
321	0.773607	0.476773
70	0.228714	0.499995
209	0.364619	0.517926
656	0.775536	0.503051
685	0.303287	0.490986

	Model	MSE
3	Gradient Boosting Regressor	0.034329
4	Mean Function	0.034627
1	Decision Tree Regressor	0.034693
2	Random Forest Regressor	0.034701
0	Linear Regressor	0.034808

	Actual	Predicted
321	0.233319	0.435826
70	0.613672	0.491103
209	0.635615	0.516908
656	0.551901	0.515739
685	0.321437	0.481831

	Model	MSE
4	Mean Function	0.033857
3	Gradient Boosting Regressor	0.034368
2	Random Forest Regressor	0.035164
1	Decision Tree Regressor	0.035352
0	Linear Regressor	0.035449

	Данные	Прогноз
0	0.487579	0.508204
1	0.292155	0.509448
2	0.245958	0.522566
3	0.850789	0.504228
4	0.611147	0.469178
...	...	...
276	0.528359	0.470481
277	0.440618	0.482079
278	0.834572	0.534503
279	0.534751	0.511791
280	0.569121	0.473284